Temporarily disabling an AI agent so it stops processing messages and executing actions, without destroying its configuration or data.
Agent suspension is a controlled pause. The agent stops responding to messages, stops executing tools, and stops consuming LLM tokens. But its configuration, workspace data, conversation history, and credentials remain intact.
You suspend an agent when something goes wrong: unexpected behavior, cost spikes, security concerns, or during maintenance. Unlike a kill switch (immediate emergency stop), suspension is a deliberate operational action.
When you are ready, resume the agent with one click. It picks up where it left off with all data intact.
You need the ability to pause agents without destroying them. Suspension lets you investigate issues, apply fixes, and resume — all without losing state or requiring reprovisioning.
Clawctl provides one-click agent suspension and resumption from the dashboard. Suspended agents consume no LLM costs. Auto-recovery is paused during suspension. All data is preserved.
Try Clawctl — 60 Second DeployNo LLM costs during suspension. The Clawctl platform fee continues as normal.
Messages receive a configurable "agent unavailable" response. They are not queued or lost.
Instant. One click restarts the agent. It is available within seconds.
Kill Switch
An emergency mechanism that immediately stops an AI agent from taking any further actions when triggered by an operator.
Agent Monitoring
Real-time observation of AI agent behavior, performance, and health — including conversation quality, error rates, and resource usage.
Agent Recovery
Automated detection and correction of agent failures — including container crashes, health check failures, and degraded performance.
Health Checks
Automated probes that verify an AI agent is running, responsive, and functioning correctly at regular intervals.